Dialogue is a communicative activity based on verbal/nonverbal interaction. Research on dialogue has been conducted in various disciplines to examine various contexts, production and reception mechanisms, dialogic structures as well as varied functions and implications of dialogic interaction. Conversational AI is an umbrella term used to describe various methods of enabling computers to carry on a conversation with a human. This technology ranges from fairly simple natural language processing (NLP) applications to more sophisticated machine learning (ML) and Large Language Models (LLMs) that can interpret a wider range of inputs and conduct more complex conversations. The study attempts a contrastive analysis of three authentic human-human conversations and their human-bot simulations by character.ai (https://character.ai) chatbots to assess whether the virtual interactions succeed in simulating the real interactive experience. Pickering and Garrod's (2021) Interactive Alignment model is adopted as the framework of the analysis. Selected NLP applications are employed to assess lexical and semantic alignment in addition to other dialogue production mechanisms, namely routinization and AI hallucination. The results show that the human-bot interactions provide successful simulations of the human-human interaction; however, some human interactional features are still missing. Fine-tuning of the language model employed by the bot is recommended to maximize the authenticity of the interactive experience.